Stochastic inventory systems in a supply chain with asymmetric information: Cycle stocks, safety stocks, and consignment stock

The two critical factors distinguishing inventory management in a multifirm supply-chain context from the more traditional centrally planned perspective are incentive conflicts and information asymmetries. We study the well-known order quantity/reorder point (Q, r) model in a two-player context, using a framework inspired by observations during a case study. We show how traditional allocations of decision rights to supplier and buyer lead to inefficient outcomes, and we use principal-agent models to study the effects of information asymmetries about setup cost and backorder cost, respectively. We analyze two "opposite" models of contracting on inventory policies. First, we derive the buyer's optimal menu of contracts when the supplier has private information about setup cost, and we show how consignment stock can help reduce the impact of this information asymmetry. Next, we study consignment and assume the supplier cannot observe the buyer's backorder cost. We derive the supplier's optimal menu of contracts on consigned stock level and show that in this case, the supplier effectively has to overcompensate the buyer for the cost of each stockout. Our theoretical analysis and the case study suggest that consignment stock helps reduce cycle stock by providing the supplier with an additional incentive to decrease batch size, but simultaneously gives the buyer an incentive to increase safety stock by exaggerating backorder costs. This framework immediately points to practical recommendations on how supply-chain incentives should be realigned to overcome existing information asymmetries.